«by Dyron Keith Dabney A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy (Political Science) ...»
principles and philosophy was a creed that paid homage to a “sunao” (unbiased and humbled) mind and spirit.16 Now in its twenty-seventh year since it creation, the Matsushita Institute is noted for producing politicians traditionally associated with Japan’s elite universities like Waseda, Keio and Tokyo. In the beginning, however, MIGM drew public suspicion when it admitted its first students (called “associates”) in 1981 because of the unconventional mission of the program and highly selective admission process—only 20 students were admitted in the first entering class. Twelve years later in1993, MIGM had come of age.
In 1993, the political conditions were ideal for a number of MIGM alums to run for the Lower House. Backed by the name value of MIGM and mostly endorsed by the new political parties, 23 MIGM graduates contested Lower House seats in the 1993 general election. Prior to the 1993 Lower House election only one MIGM graduate, Aisawa Ichiro (House of Representatives, Okayama 1st district) held a seat in Diet. After the election, the total numbered fifteen.17 When these graduates collectively emerged in 1993 as Lower House candidates, they caused quite a public stir. Suddenly MIGM was the source of media attention since its graduates showcased their MIGM affiliation as much as their political party affiliation. Observers speculated whether the MIGM would 16 A recitation of the creed took place each morning as a prologue to each work day by students and the
administrative staff in attendance at the Institute. Translated from Japanese, the creed reads as follows:
“With a sunao mind, we firmly dedicate ourselves to gathering wisdom, seeking out the intrinsic nature of reality through independent study, and searching anew every day for the path that will lead to new growth and development. With deep love for our country and our people, we seek to contribute to the peace, happiness and prosperity of all humankind by searching or guiding principles of government and management based on a new vision of the nature of human beings.” 17 Presently, 70 of the 230 graduates of MIGM occupy local and national assembly seats throughout Japan.
The five-year program has since been cut to 3 years.
Clean Government Party —its graduates running in 1993 certainly outnumbered the candidates backed by smaller opposition parties. The declared political party affiliations of these candidates, however, put such speculation to rest. MIGM was more keen on graduating quality candidates for public office than new political party building. Besides, the new political parties in 1993 (JNP, JRP and Sakigake) had sufficiently filled the progressive-conservative gap in the ideological continuum between Japanese conservative parties and liberal parties.
Case Studies of Three MIGM Graduates As MIGM graduates running for the Lower House for the first time under new political party labels, the three candidates are quite similar with the exception of the district type—urban, suburban and rural—in which they competed, and in their degrees of success. The districts are fairly close to each other in the Kanto and southern Tohoku areas of Japan. The urban candidate is Usami Noboru. Usami competed in the Tokyo 2nd, 3rd and 4th districts in the 1993, 1996 and 2000 Lower House elections, respectively.
He won is first election and lost the next two. The suburban candidate is Matsuzawa Shigefumi. Matsuzawa competed in the Kanagawa 2nd district in 1993 and the Kanagawa 9th district in 1996 and 2000, respectively, and won all of his elections convincingly.
Finally, the rural candidate is Genba Kouichirou. Genba competed in the Fukushima 2nd district in 1993 and Fukushima 3rd district in 1996 and 2000. He won the first election, barely lost the second (but was elected on the PR list), and regained his SMD seat in the third. The political party affiliations of these candidates were all opposition parties that
many other details are in the case studies reported in Chapters 3 and 4.
While the study analyzes campaign behavior of three candidates, the suburban candidate, Matsuzawa, has been given a chapter of his own and is treated at more length.
The additional attention to Matsuzawa’s campaign behavior partly is to address an underrepresentation of political science scholarship on suburban electoral politics in Japan compared to urban and rural. Under the former multi-member district (MMD) system, one-third of the 511 MMD seats were regarded as suburban. Under the new system, a slightly smaller proportion (28 percent) of the 300 SMD seats as appears to be suburban.18 Conclusions reached about electioneering in the suburban district typically are extrapolations of rural and urban districts campaign data. Hence, a focus on Matsuzawa’s campaign behavior represents a focus on the electoral challenges particular to the suburban district. Indeed, all Lower House candidates—rural, suburban and urban—face some constituency population fluctuations that may influence voting patterns, while the voter behavior for the rural and urban districts has remained the relatively stable and consistent. Yet, the suburban electorate from which Matsuzawa sought support reflects an influx of individuals from rural and urban locations who bring with them the respective political temperaments of urban and rural districts. The diversity of the socio-economic status, age, occupation, and political attitudes of the electorate in suburban districts, and the continual influx of new residents through urban 18 These figures were derived by tabulating the total number of urban/metropolitan, mixed (suburban) and rural districts listed in Steven Reed’s source book, Japan Election Data: The House of Representative, 1947-1990 and Seiji Handbook (No. 30, p.195). Reed adopts the Asahi Shimbun’s demographic/geographic categorizations for election districts.
behavior patterns (Ben-Ari, 1991). Consequently, suburban election districts, due their socio-demographic diversity, suggest less voter stability and predictability.
With regard to the three elections included in this dissertation, each has its own particular interest, with one pre-reform and two post-reform. There is an excellent book mostly written by Steven R. Reed (2003), Japanese Electoral Politics: Creating a New Party System, that also focuses on the same three elections, including three chapters (written by others) on particular regions. This book clearly demonstrates how much electoral dynamics shifted across the three elections, but its focus on the political party system actually says less than one might expect about changes (or non-changes) in election campaign strategies.
In brief, the conventional wisdom indicates that we should have witnessed more Type B campaign tactics in the nation as a whole in the 1993 Lower House election, given the three new parties (two split off from the LDP) and the incessant calls for reform. That should particularly be the case with our candidates, as first-time challengers and recent alums of MIGM. On the other hand, the 1993 election was still conducted under the MMD system with all its incentives for Type A campaign tactics. In 1996, turmoil continued, and the reform had been carried out, again presumably favoring the Type B strategy. Most of the extensive coverage at the time (see Chapter 2) expressed disappointment that this did not happen (see the concluding chapter for some reasons why not). Finally, the 2000 election was for several reasons a more “normal” case to look at the effects of reform than was its first example, and our study is distinctive in including that perspective.
looks at Matsuzawa’s three campaigns in his suburban district, and then a chapter that does the same in a briefer form for urban candidate, Usami and rural candidate, Genba.
These chapters are essentially descriptive case studies aimed more at presenting evidence than analyzing it. The conclusion will propose some hypotheses to explain the variations we observed among the nine campaigns. While these propositions cannot be rigorously tested for applicability to national patterns, given the small sample, they do lead to some interesting and new implications.
The underlying question in this dissertation is the point that puzzles most people who look at Japanese elections—why do they still look so “traditional,” or in our terms, so favor the Type A strategy? Why did the prediction by Curtis cited at the beginning of this chapter not come true? We do not have an authoritative answer, but our interpretations based on detailed participant-observation research throw some new light on the subject.
Gerald Curtis’ best-selling book Daigishi no Tanjou (Birth of a Dietman), published in 1971, was an authoritative analysis of campaign behavior in Japan, and even today is often referenced by Japanese politicians and academics alike.19 His research was a detailed case study of one rural campaign in the 1967 Lower House election that codified how a candidate navigated the MMD election system and the labyrinth of election campaign rules and restrictions uncommon among representative democracies.20 Corresponding to the list of dichotomies presented in the introductory chapter, Curtis’s accounts of the “hard vote” and the “gathered vote” on the left side, “floating voters” on the right side, and kouenkai or support associations as a then-recent innovation spanning both sides, provided the context for understanding how Japanese election campaigns work.
Explanations for Japan’s Distinctive Campaign Behavior Although Curtis’s work was not explicitly comparative with respect to district types in Japan (he examines only one political candidate), or across other democratic 19 The English version, Election Campaigning Japanese Style (also 1971) played a similarly seminal role for the far smaller number of foreign academics who study Japanese politics.
20 With this exception, this review concentrates on writings in English by political scientists. The literature written in Japanese on elections is vast, beyond our ability to treat comprehensively. Literature specific to candidates’ campaign behavior, however, is quite limited. Election campaigning per se has been more a topic for journalism than for academic political science in Japan; we believe that most of the important analysis on this subject can be found in the works cited here.
quite distinctive. Curtis stressed two explanations for the observed distinctiveness of Japanese election campaigns: 1) restrictive campaign rules; and 2) the election system.21 Japanese election campaigns continue to be governed by highly restrictive rules.
In fact, the number of rules has only increased over the years. Candidates are prohibited from using advertising even in newspapers, much less radio and television, and door-todoor canvassing by campaign workers or volunteers. The number of formal public speeches, posters, and sound-trucks is limited, and voters cannot be offered as much as a cup of tea or coffee if they visit a campaign office. How money can be raised and spent is tightly regulated.
Three reasons have been advanced for such restrictive rules. One is to give a reasonable chance to candidates who do not have much money. Another is to prevent gathering votes by “traditional” inducements such as reciprocity for some gift or favor, or simply direct requests from a respected person. A third can be inferred from the fact that many specific restrictive practices dated from 1925, the year that universal manhood suffrage was instituted. With an expanded, but less informed electorate in Japan, elite political “caretakers” at the time wanted to inhibit the ability of “demagogues” to whip up public emotion, and later elites left these provisions in place presumably for similar reasons.
21 Two parallel laws govern Japan’s election campaigns. The Public Offices Elections Law (POEL) regulates the rules of the game of Japanese elections ranging from campaign activities to district magnitude, and the Political Funds Control Law (PFCL), defines campaign spending (seiji kenkin) limits.
On the importance of electoral laws in general, see Saito, 1995; and Grofman and Lijphard 1986. For an account of these provisions in Japan, including recent changes, see Jain, 1993; Seligmann, 1997; Otake, et.
al., 1998; Christensen, 1996 and 1998, and Reed, 1999.
candidates are simply not allowed to do much for gathering votes and mobilizing voters—a luxury taken for granted by candidates in other democracies. Incidentally, these restrictions have probably had the net effect of making campaigns more rather than less “traditional” given that many “modern” campaign tactics were not sanctioned under Japan’s Political Funds Control Law, which places limits on campaign spending (seiji kenkin). It remains relatively easy to regulate campaign behaviors, including most ways money is spent during the campaign. However, it is harder to regulate how money is raised and regulate other activities that can be carried out behind closed doors that invite campaign corruption. Disclosures of campaign spending usually have been far from accurate since political parties and candidates have been particularly savvy about taking advantage of the loopholes on financial reports. In fact, Japanese election campaigns require vast sums of money, far more than could be spent legally. In the twelve days of the official election campaign in 1996 alone, for example, candidates’ averaged campaign expenses were well over fifty percent of the 24 million yen spending limit for the SMDs.
Yet, the ceiling on spending typically was set high enough that candidates did not exceed the spending limit (Carlson, 2007). It is assumed that much of the money is spent, in effect, buying votes, albeit indirectly by providing cash to influential people to obtain their support, then providing them with the means to encourage other to follow suit, rather than through direct payments to voters.22 Herein rests the opportunity for corruption to enter the electoral and political processes.